pnnl/EXPERT2

RESOURCE

Abstract

This software includes the Jupyter notebooks, model pretraining and evaluation code for the EXPERT 2.0 Human-AI Reasoning Engine V0.1. It supports pre-training a Human-AI model for reasoning over multi-layer network representations. It includes prompt-based evaluation framework in a Jupyter notebook for AI reasoning and Jupyter widgets with AI-based techniques for evidence generation and uncertainty quantification to support human-AI reasoning.
Developers:
Horawalavithana, Sameera [1] Munikoti, Sai [2] Wagle, Sridevi [2] Acharya, Anurag [2] Sharma, Shivam [2]
  1. Pacific Northwest National Laboratory
  2. Pacific Northwest National Laboratory (PNNL), Richland, WA (United States)
Release Date:
2023-10-18
Project Type:
Open Source, Publicly Available Repository
Software Type:
Scientific
Licenses:
BSD 2-clause "Simplified" License
Sponsoring Org.:
Code ID:
114736
Site Accession Number:
Battelle IPID 32848-E
Research Org.:
Pacific Northwest National Laboratory (PNNL), Richland, WA (United States)
Country of Origin:
United States

RESOURCE

Citation Formats

Horawalavithana, Sameera, Munikoti, Sai, Wagle, Sridevi, Acharya, Anurag, and Sharma, Shivam. pnnl/EXPERT2. Computer Software. https://github.com/pnnl/EXPERT2. USDOE. 18 Oct. 2023. Web. doi:10.11578/dc.20231018.1.
Horawalavithana, Sameera, Munikoti, Sai, Wagle, Sridevi, Acharya, Anurag, & Sharma, Shivam. (2023, October 18). pnnl/EXPERT2. [Computer software]. https://github.com/pnnl/EXPERT2. https://doi.org/10.11578/dc.20231018.1.
Horawalavithana, Sameera, Munikoti, Sai, Wagle, Sridevi, Acharya, Anurag, and Sharma, Shivam. "pnnl/EXPERT2." Computer software. October 18, 2023. https://github.com/pnnl/EXPERT2. https://doi.org/10.11578/dc.20231018.1.
@misc{ doecode_114736,
title = {pnnl/EXPERT2},
author = {Horawalavithana, Sameera and Munikoti, Sai and Wagle, Sridevi and Acharya, Anurag and Sharma, Shivam},
abstractNote = {This software includes the Jupyter notebooks, model pretraining and evaluation code for the EXPERT 2.0 Human-AI Reasoning Engine V0.1. It supports pre-training a Human-AI model for reasoning over multi-layer network representations. It includes prompt-based evaluation framework in a Jupyter notebook for AI reasoning and Jupyter widgets with AI-based techniques for evidence generation and uncertainty quantification to support human-AI reasoning.},
doi = {10.11578/dc.20231018.1},
url = {https://doi.org/10.11578/dc.20231018.1},
howpublished = {[Computer Software] \url{https://doi.org/10.11578/dc.20231018.1}},
year = {2023},
month = {oct}
}